Machine learning and radiomics for ventricular tachyarrhythmia prediction in hypertrophic cardiomyopathy: insights from an MRI-based analysis.

BACKGROUND: Myocardial fibrosis is often detected in patients with hypertrophic cardiomyopathy (HCM), which causes left ventricular (LV) dysfunction and tachyarrhythmias. PURPOSE: To evaluate the potential value of a machine learning (ML) approach that uses radiomic features from late gadolinium enhancement (LGE) and cine images for the prediction of ventricular tachyarrhythmia (VT) in patients with HCM. MATERIAL AND METHODS: Hyperenhancing areas of LV myocardium on LGE images were manually segmented, and the segmentation was propagated to corresponding areas on cine images. Radiomic features were extracted using the PyRadiomics library. The least absolute shrinkage and selection operator (LASSO) method was employed for radiomic feature selection. Our model development employed the TabPFN algorithm, an adapted Prior-Data Fitted Network design. Model performance was evaluated graphically and numerically over five-repeat fivefold cross-validation. SHapley Additive exPlanations (SHAP) were employed to determine the relative importance of selected radiomic features. RESULTS: Our cohort consisted of 60 patients with HCM (73.3% male; median age = 51.5 years), among whom 17 had documented VT during the follow-up. A total of 1612 radiomic features were extracted for each patient. The LASSO algorithm led to a final selection of 18 radiomic features. The model achieved a mean area under the receiver operating characteristic curve of 0.877, demonstrating good discrimination, and a mean Brier score of 0.119, demonstrating good calibration. CONCLUSION: Radiomics-based ML models are promising for predicting VT in patients with HCM during the follow-up period. Developing predictive models as clinically useful decision-making tools may significantly improve risk assessment and prognosis.

Investigating the relationship between cognitive impairment and brain white matter tracts using diffusion tensor imaging in patients with prolactinoma.

BACKGROUND: Cognitive impairment is known to occur in patients with prolactinoma, but the underlying mechanism is unclear. OBJECTIVE: To evaluate cognitive function in patients with prolactinoma and to investigate the basis of possible cognitive impairment in brain white matter changes using diffusion tensor imaging (DTI). METHODS: 37 consecutive patients with prolactinoma and 37 healthy controls of similar age, sex, and education were enrolled in the study. Hormone levels were determined in all participants, comprehensive neuropsychological testing was performed, and DTI was used to reconstruct and evaluate white matter tracts. RESULTS: In patients with prolactinoma, short- and long-term visual and verbal memory, attention, concentration, and executive and language functions were impaired compared to the healthy group. When comparing the DTI results, lower fractional anisotropy (FA) values were found in the patients' right uncinate fasciculus (R-UF), indicating neuronal damage. After applying the Bonferroni correction, the two groups had no significant difference in 42 tracts (p > 0.0012 for all). A positive correlation was found between poor FA scores on the R-UF and low scores on long-term memory, category and letter fluency tests. In addition, patients with hypoprolactinemia had the worst short-term memory scores, while normoprolactinemia had the best scores. Also, the poorer R-UF FA values were found in the patients with hypoprolactinemia and the highest in those with normoprolactinemia. CONCLUSION: This study is the first to investigate reasons for cognitive dysfunction in patients with prolactinoma by DTI. No significant structural changes were found in brain tracts of patients with prolactinoma. Still, there may be a link between potential damage in the R-UF and cognitive dysfunction, and further research is needed. In addition, the results showed that the development of hypoprolactinemia is associated with cognitive dysfunction and emphasized that overtreatment should be avoided.

Brain white matter microstructural alterations in Behcet's syndrome correlate with cognitive impairment and disease severity: A diffusion tensor imaging study.

OBJECTIVES: To evaluate the microstructural integrity of brain white matter tracts in patients with Neuro-Behcet's syndrome (NBS) and Behcet's syndrome (BS) without neurological manifestations using diffusion tensor imaging (DTI) and to investigate potential utility of DTI as a surrogate biomarker of neurocognitive functioning and disease severity. METHODS: This cross-sectional study comprised 34 NBS patients and 32 BS patients without neurological involvement, identified based on the International Study Group of the Behcet's disease (ISGBD) and the International Consensus Recommendation (ICR) criteria, as well as 33 healthy controls. Cognitive functions, including attention, memory, language, abstraction, executive control, visuospatial skills, and sensorimotor performance were assessed using standardized questionnaires. DTI data were analyzed using tract-based spatial statistics (TBSS) and automated probabilistic tractography to investigate inter-group differences. Subsequently, correlations between tensor-derived parameters of white matter tracts, neurocognitive test scores, and disease severity measures were examined. RESULTS: DTI revealed decreased fractional anisotropy and increased radial diffusivity, mean diffusivity, and axial diffusivity in both supratentorial and infratentorial white matter in NBS patients, indicating widespread loss of microstructural integrity. Moreover, this loss of integrity was also observed in BS patients without neurological manifestations, albeit to a lesser extent. In NBS patients, certain white matter tracts, including cingulum bundle, were associated with poor cognitive performance across multiple domains and disease severity. DISCUSSION: DTI findings might potentially serve as a neuroimaging marker to predict the extent of neurocognitive impairment and disease severity associated with central nervous system involvement in BS.

Structural and Functional Impact of Adrenoceptor Beta-1 Gene Polymorphism in Patients with Hypertrophic Cardiomyopathy and Response to Beta-Blocker Therapy.

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is a genetically inherited cardiac disorder with diverse clinical presentations. Adrenergic activity, primarily mediated through beta-adrenoceptors, plays a central role in the clinical course of HCM. Adrenergic stimulation increases cardiac contractility and heart rate through beta-1 adrenoceptor activation. Beta-blocker drugs are recommended as the primary treatment for symptomatic HCM patients to mitigate these effects. METHODS: This prospective study aimed to investigate the impact of common ADRB-1 gene polymorphisms, specifically serine-glycine at position 49 and arginine-glycine at position 389, on the clinical and structural aspects of HCM. Additionally, the study explored the association between these genetic variations and the response to beta-blocker therapy in HCM patients. RESULTS: A cohort of 147 HCM patients was enrolled, and comprehensive assessments were performed. The findings revealed that the Ser49Gly polymorphism significantly influenced ventricular ectopic beats, with beta-blocker therapy effectively reducing them in Ser49 homozygous patients. Moreover, natriuretic peptide levels decreased, particularly in Ser49 homozygotes, indicating improved cardiac function. Left ventricular outflow obstruction, a hallmark of HCM, was also reduced following beta-blocker treatment in all patient groups. In contrast, the Arg389Gly polymorphism did not significantly impact baseline parameters or beta-blocker response. CONCLUSION: These results emphasize the role of the Ser49Gly polymorphism in the ADRB-1 gene in shaping the clinical course and response to beta-blocker therapy in HCM patients. This insight may enable a more personalized approach to managing HCM by considering genetic factors in treatment decisions. Further research with larger populations and longer follow-up periods is needed to confirm and expand upon these findings.

Pulmonary Hypertension Due to High-Output Heart Failure: Hereditary Hemorrhagic Telangiectasia.

Pulmonary hypertension (PH) is a complex disorder that should be managed with a multidisciplinary approach. Although most of the underlying causes of left heart disease can be easily diagnosed with cardiac imaging, some pathologies might necessitate careful investigation to go beyond the obvious. High-output heart failure (HF) due to arteriovenous malformation (AVMs) is an unnoticeable cause for HF and PH. Patients with hepatic AVMs should always be carefully evaluated with regard to hereditary hemorrhagic telangiectasia (HHT) since they can have multiple signs related to the other systems without any symptoms. In this case report, we discussed a patient who was initially diagnosed as PH associated with HF with preserved ejection fraction but eventually was found to have PH associated with high-output HF due to hereditary hemorrhagic telangiectasia (HHT, or Osler Weber Rendu syndrome) after detailed evaluation.

Can diffusion tensor imaging have a diagnostic utility to differentiate early-onset forms of bipolar disorder and schizophrenia: A neuroimaging study with explainable machine learning algorithms.

BACKGROUND/AIM: Accurate diagnosis of early-onset psychotic disorders is crucial to improve clinical outcomes. This study aimed to differentiate patients with early-onset schizophrenia (EOS) from early-onset bipolar disorder (EBD) with machine learning (ML) algorithms using white matter tracts (WMT). METHOD: Diffusion tensor imaging was obtained from adolescents with either EOS (n = 43) or EBD (n = 32). Global probabilistic tractography using an automated tract-based TRACULA software was performed to analyze the fractional anisotropy (FA) of forty-two WMT. The nested cross-validation was performed in feature selection and model construction. EXtreme Gradient Boosting (XGBoost) was applied to select the features that can give the best performance in the ML model. The interpretability of the model was explored with the SHApley Additive exPlanations (SHAP). FINDINGS: The XGBoost algorithm identified nine out of the 42 major WMTs with significant predictive power. Among ML models, Support Vector Machine-Linear showed the best performance. Higher SHAP values of left acoustic radiation, bilateral anterior thalamic radiation, and the corpus callosum were associated with a higher likelihood of EOS. CONCLUSIONS: Our findings suggested that ML models based on the FA values of major WMT reconstructed by global probabilistic tractography can unveil hidden microstructural aberrations to distinguish EOS from EBD.

Altered thalamic volume in patients with mild autonomous cortisol secretion: a structural brain MRI study.

PURPOSE: To compare thalamic volume and cognitive functions of patients with mild autonomous cortisol secretion (MACS) with control subjects and patients with overt Cushing's syndrome (CS). METHODS: In this cross-sectional study, volumes of regions of interest were assessed using 3 T magnetic resonance imaging and a voxel-based morphometry approach in 23 patients with MACS, 21 patients with active CS, 27 patients with CS in remission, and 21 control subjects. Cognitive functions were assessed using validated questionnaires. RESULTS: Patients with MACS had smaller left thalamic (F = 3.8, p = 0.023), left posterior thalamic (F = 4.9, p = 0.01), left medial thalamic (F = 4.7, p = 0.028), and right lateral thalamic (F = 4.1, p = 0.025) volumes than control subjects. Patients with active CS also had smaller left thalamic (F = 3.8, p = 0.044), left posterior thalamic (F = 4.9, p = 0.007), left medial thalamic (F = 4.7, p = 0.006), and right lateral thalamic (F = 4.1, p = 0.042) volumes compared to controls. Patients with CS in remission had smaller left medial (F = 4.7, p = 0.030) and right lateral thalamic (F = 4.1, p = 0.028) volumes than controls. Neuropsychological tests showed no difference between the groups. CONCLUSION: MACS may decrease thalamic volume.

Radiomics-based machine learning models in STEMI: a promising tool for the prediction of major adverse cardiac events.

OBJECTIVE: To evaluate the potential value of the machine learning (ML) models using radiomic features of late gadolinium enhancement (LGE) and cine images on magnetic resonance imaging (MRI) along with relevant clinical information and conventional MRI parameters for the prediction of major adverse cardiac events (MACE) in ST-segment elevation myocardial infarction (STEMI) patients. METHODS: This retrospective study included 60 patients with the first STEMI. MACE consisted of new-onset congestive heart failure, ventricular arrhythmia, and cardiac death. Radiomic features were extracted from cine and LGE images. Inter-class correlation coefficients (ICCs) were calculated to assess inter-observer reproducibility. LASSO (least absolute shrinkage and selection operator) method was used for radiomic feature selection. Seven separate models using a different combination of the available information were investigated. Classifications with repeat random sampling were done using adaptive boosting, k-nearest neighbor, naive Bayes, neural network, random forest, stochastic gradient descent, and support vector machine algorithms. RESULTS: Of the 1748 extracted radiomic features, 1393 showed good inter-observer agreement. With LASSO, 25 features were selected. Among the ML algorithms, the neural network showed the highest predictive performance on average (area under the curve (AUC) 0.822 ± 0.181). Of the best-calculated model, the one using clinical parameters, CMRI parameters, and selected radiomic features (model 7), the diagnostic performance was as follows: 0.965 AUC, 0.894 classification accuracy, 0.906 sensitivity, 0.883 specificity, 0.875 positive predictive value (PPV), and 0.912 negative predictive value (NPV). CONCLUSION: The radiomics-based ML models incorporating clinical and conventional MRI parameters are promising for predicting MACE occurrence in STEMI patients in the follow-up period. KEY POINTS: • Acute coronary occlusion results in variable changes at the cellular level ranging from myocyte swelling to myonecrosis depending on the duration of the ischemia and the metabolic state of the heart, which causes subtle heterogeneous signal changes that are imperceptible to the human eye with cardiac MRI. • Radiomics-based machine learning analysis of cardiac MR images is promising for risk prediction. • Combining MRI-derived parameters and clinical variables increases the accuracy of predictive models.

A Rare Cause of Breast Swelling in Lactating Women: Aquafilling® Gel Injection.

Aquafilling® gel has been used in recent years as an alternative method to breast augmentation surgery. In this case report, we aimed to discuss radiological imaging findings and complications of Aquafilling® gel injection. Case: A 34-year-old lactating female patient presented with swelling and pain in the left breast. Ultrasonography showed massive septate fluid collections with dense content in both breasts and between the pectoral muscle fibers. On mammography, both breasts appeared dense with large mass opacities. On magnetic resonance imaging, extensive fluid-signal cystic areas were observed. Peripheral enhancement around the fluid in the left breast was present in the contrast-enhanced series. We learned from her anamnesis that Aquafilling® was applied to both breasts 5 years ago. Ultrasonography-guided sampling was performed from the cystic areas in the left breast and cytological examination revealed basophilic Aquafilling® material surrounded by diffuse inflammatory infiltrates. Breast augmentation history should be questioned in suspicious cases because Aquafilling® gel injection and its complications can present with a wide variety of symptoms and may mimic various other conditions on radiological imaging, such as cancer, abscess, granulomatous mastitis, and parasite infections.

Potential Association Between COVID-19 Vaccination and Facial Palsy: Three Cases With Neuroimaging Findings.

Acute onset Facial palsy was reported in four vaccinated participants in the BNT162b2 (Pfizer-BioNTech) vaccine clinical trials published on December 10, 2020. So far, few cases of Facial palsy among the mRNA vaccine groups have been previously documented in the literature. Facial palsy is cited as medically attended adverse event following immunization on April 12, 2021, after the first dose of the approved Pfizer-BioNTech COVID-19 vaccines for preventive immunization for SARS-CoV-2 is administrated to the population in Turkey. This study is aimed to describe clinical and magnetic resonance imaging features of three patients, who developed acute onset peripheral facial paralysis after administration of the BNT162b2 vaccine, without any previous medical condition. The first patient presented with right sided facial palsy within the same day following the vaccine was administrated, while the second patient presented with left sided facial palsy 2 months after vaccination. The third patient, on the other hand, presented with right sided facial palsy and abducens nerve (CN VI) paralysis two days after vaccine was administrated.

Plasma Viral miRNAs Indicate a High Prevalence of Occult Viral Infections.

Prevalence of Kaposi sarcoma-associated herpesvirus (KSHV/HHV-8) varies greatly in different populations. We hypothesized that the actual prevalence of KSHV/HHV8 infection in humans is underestimated by the currently available serological tests. We analyzed four independent patient cohorts with post-surgical or post-chemotherapy sepsis, chronic lymphocytic leukemia and post-surgical patients with abdominal surgical interventions. Levels of specific KSHV-encoded miRNAs were measured by reverse transcription-quantitative polymerase chain reaction (RT-qPCR), and KSHV/HHV-8 IgG were measured by immunoassay. We also measured specific miRNAs from Epstein Barr Virus (EBV), a virus closely related to KSHV/HHV-8, and determined the EBV serological status by ELISA for Epstein-Barr nuclear antigen 1 (EBNA-1) IgG. Finally, we identified the viral miRNAs by in situ hybridization (ISH) in bone marrow cells. In training/validation settings using independent multi-institutional cohorts of 300 plasma samples, we identified in 78.50% of the samples detectable expression of at least one of the three tested KSHV-miRNAs by RT-qPCR, while only 27.57% of samples were found to be seropositive for KSHV/HHV-8 IgG (P<0.001). The prevalence of KSHV infection based on miRNAs qPCR is significantly higher than the prevalence determined by seropositivity, and this is more obvious for immuno-depressed patients. Plasma viral miRNAs quantification proved that EBV infection is ubiquitous. Measurement of viral miRNAs by qPCR has the potential to become the "gold" standard method to detect certain viral infections in clinical practice.